HPML projects multi-agent update fields onto the closest metric-gradient potential flow via Hodge decomposition, yielding Lyapunov potentials and equilibrium-gap bounds.
A variational inequality perspective on generative adversarial networks
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Scion is a new stochastic LMO-based optimizer family that unifies existing methods, supports unconstrained problems, and delivers hyperparameter transferability plus speedups on nanoGPT training.
SGD is reformulated via a master equation from discrete updates, producing a discrete Fokker-Planck equation that predicts non-stationary variance growth proportional to learning rate in flat Hessian directions.
Mirror descent algorithms with productive/non-productive step switching achieve optimal convergence rates for bounded monotone operators and Lipschitz convex functional constraints in variational inequalities.
Presents a stochastic extragradient algorithm for variational inequalities with Markovian noise, proving convergence under L-Lipschitzness, strong monotonicity, and noise bounded only at the optimum, plus experiments on mixing time.
citing papers explorer
-
Metric-Gradient Projection for Stable Multi-Agent Policy Learning
HPML projects multi-agent update fields onto the closest metric-gradient potential flow via Hodge decomposition, yielding Lyapunov potentials and equilibrium-gap bounds.
-
Training Deep Learning Models with Norm-Constrained LMOs
Scion is a new stochastic LMO-based optimizer family that unifies existing methods, supports unconstrained problems, and delivers hyperparameter transferability plus speedups on nanoGPT training.
-
Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics
SGD is reformulated via a master equation from discrete updates, producing a discrete Fokker-Planck equation that predicts non-stationary variance growth proportional to learning rate in flat Hessian directions.
-
Mirror Descent-Type Algorithms for the Variational Inequality Problem with Functional Constraints
Mirror descent algorithms with productive/non-productive step switching achieve optimal convergence rates for bounded monotone operators and Lipschitz convex functional constraints in variational inequalities.
-
Methods for Solving Variational Inequalities with Markovian Stochasticity
Presents a stochastic extragradient algorithm for variational inequalities with Markovian noise, proving convergence under L-Lipschitzness, strong monotonicity, and noise bounded only at the optimum, plus experiments on mixing time.
- A path-following framework on fiber bundle for variational inequalities